Update src/streamlit_app.py

src/streamlit_app.py

@@ -2,24 +2,13 @@ import os
 import numpy as np
 import tensorflow as tf
 from PIL import Image
-import logging
 from pathlib import Path
-# Note: ConvNeXtLarge weights='imagenet' will be downloaded by TensorFlow,
-# which also has its own caching mechanism. This is separate from the
-# huggingface_hub download for the .h5 model weights.
 from tensorflow.keras.applications import ConvNeXtLarge
 import streamlit as st
 import io
 from huggingface_hub import hf_hub_download, try_to_load_from_cache
 
-# Configure logging
-setup_logging()
-
-def setup_logging():
-    logging.basicConfig(
-        level=logging.INFO,
-        format='%(asctime)s - %(levelname)s - %(message)s'
-    )
+# Removed: setup_logging function and its call
 
 # We will now rely on hf_hub_download's default cache behavior
 def download_model_from_hub():
@@ -82,16 +71,16 @@ def load_model(model_path):
     Loads the Keras model weights from a specified path.
     """
     try:
-        logging.info(f"Attempting to load model from {model_path}")
+        # Removed: logging.info(f"Attempting to load model from {model_path}")
         # First create the model architecture
         model = create_convnext_model()
 
         # Then load the weights from the downloaded .h5 file
         model.load_weights(model_path)
-        logging.info("Model weights loaded successfully.")
+        # Removed: logging.info("Model weights loaded successfully.")
         return model
     except Exception as e:
-        logging.error(f"Error loading model weights: {str(e)}")
+        # Removed: logging.error(f"Error loading model weights: {str(e)}")
         st.error(f"Error loading model weights: {str(e)}")
         st.info("Ensure the downloaded file is a valid Keras .h5 weights file compatible with the ConvNeXtLarge architecture.")
         # Re-raise the exception so Streamlit knows to stop if model loading fails
@@ -109,7 +98,7 @@ def preprocess_image(image, target_size=(512, 512)):
         # Resize image using BICUBIC for potentially better quality than BILINEAR for downsampling
         img = image.resize(target_size, Image.Resampling.BICUBIC)
 
-        # Convert to numpy array
+        # Convert to numpy array and normalize
         img_array = np.array(img, dtype=np.float32) # Use float32 for normalization
 
         # Normalize to [0, 1]
@@ -129,13 +118,14 @@ def preprocess_image(image, target_size=(512, 512)):
         # Add batch dimension
         img_array = np.expand_dims(img_array, axis=0)
 
-        logging.info("Image preprocessed successfully.")
+        # Removed: logging.info("Image preprocessed successfully.")
         return img_array
     except Exception as e:
-        logging.error(f"Error preprocessing image: {str(e)}")
+        # Removed: logging.error(f"Error preprocessing image: {str(e)}")
         st.error(f"Error preprocessing image: {str(e)}")
         raise
 
+
 def predict_volcano(model, image):
     """
     Makes a prediction using the loaded model.
@@ -159,7 +149,7 @@ def predict_volcano(model, image):
         # Confidence is the probability for the predicted class
         confidence = probability if probability > 0.5 else 1 - probability
 
-        logging.info(f"Prediction made: Result={result}, Probability={probability:.4f}")
+        # Removed: logging.info(f"Prediction made: Result={result}, Probability={probability:.4f}")
 
         return {
             "result": result,
@@ -167,7 +157,7 @@ def predict_volcano(model, image):
             "probability": float(probability) # Convert to standard Python float
         }
     except Exception as e:
-        logging.error(f"Error making prediction: {str(e)}")
+        # Removed: logging.error(f"Error making prediction: {str(e)}")
         st.error(f"Error making prediction: {str(e)}")
         raise
 
@@ -205,7 +195,6 @@ def get_sample_images():
 
 
 def main():
-    setup_logging()
 
     st.set_page_config(
         page_title="Volcano Detection",
@@ -249,22 +238,23 @@ def main():
         if uploaded_file is not None:
             try:
                 image = Image.open(uploaded_file)
-                logging.info("Image uploaded by user.")
+                # Removed: logging.info("Image uploaded by user.")
             except Exception as e:
                  st.error(f"Error opening uploaded image: {str(e)}")
-                 logging.error(f"Error opening uploaded image: {str(e)}")
+                 # Removed: logging.error(f"Error opening uploaded image: {str(e)}")
     elif selected_option != "Select an image" and sample_images[selected_option] is not None:
         try:
             sample_image_path = sample_images[selected_option]
             if Path(sample_image_path).exists():
                 image = Image.open(sample_image_path)
-                logging.info(f"Sample image '{selected_option}' loaded.")
+                # Removed: logging.info(f"Sample image '{selected_option}' loaded.")
             else:
                  st.warning(f"Sample image file not found: {sample_image_path}")
-                 logging.warning(f"Sample image file not found: {sample_image_path}")
+                 # Removed: logging.warning(f"Sample image file not found: {sample_image_path}")
         except Exception as e:
             st.error(f"Error loading sample image '{selected_option}': {str(e)}")
-            logging.error(f"Error loading sample image '{selected_option}': {str(e)}")
+            # Removed: logging.error(f"Error loading sample image '{selected_option}': {str(e)}")
+
 
     if image is not None:
         # Display the image